Feed aggregator

For the past few months, I've been working on a WebAssembly (WASM) component model registry and runtime (built on wasmtime) called asterai. My goal is to help make the WASM component model mainstream, because I think it's a great way to build software. I think the ecosystem is missing a few key things and an open, central registry is one of those things.

Recently I saw how ClawHub had "341 malicious skills", and couldn't help but think how WASM/WASI resolves most of these issues by default, since everything is sandboxed.

So I've spent my weekend building Asterbot, a modular AI agent where every capability is a swappable WASM component.

Want to add web search? That's just another WASM component. Memory? another component. LLM provider? component.

The components are all sandboxed, they only have access to what you explicitly grant, e.g. a single directory like ~/.asterbot (the default). It can't read any other part of the system.

Components are written in any language (Rust, Go, Python, JS), sandboxed via WASI, and pulled from the asterai registry. Publish a component, set an env var to authorise it as a tool, and asterbot discovers and calls it automatically. Asterai provides a lightweight runtime on top of wasmtime that makes it possible to bundle components, configure env vars, and run it.

It's still a proof of concept, but I've tested all functionality in the repo and I'm happy with how it's shaping up.

Happy to answer any questions!

Comments URL: https://news.ycombinator.com/item?id=46937297

Points: 1

# Comments: 0

I built this based on Karpathy's x post from a couple months ago while I was on paternity leave. It's my first product. It's using convex, openrouter, vercel workflows to orchestrate the LLMs.

Prompt multiple LLMs at once, they blindly rank each response and synthesize a final output. Saves time switching back and forth between providers and comparing responses.

In my experience, the synthesized output tends to be more reliable than any single provider

My wife and kid think it's cool. I hope other people find it useful. paid tier lets you use openrouter bring your own key, search and file attachments.

I know this requires a sign up but I didn't know how to let people try it without some way to limit usage.

https://x.com/karpathy/status/1992381094667411768?lang=en

Comments URL: https://news.ycombinator.com/item?id=46937295

Points: 1

# Comments: 0

Article URL: https://github.com/sendrec/sendrec

Comments URL: https://news.ycombinator.com/item?id=46937289

Points: 2

# Comments: 1

Article URL: https://www.latent.space/p/adversarial-reasoning

Comments URL: https://news.ycombinator.com/item?id=46936920

Points: 1

# Comments: 0

Article URL: https://angiejones.tech/stop-closing-the-door-fix-the-house/

Comments URL: https://news.ycombinator.com/item?id=46936917

Points: 1

# Comments: 0

Article URL: https://marketplace.visualstudio.com/items?itemName=Hud.hud

Comments URL: https://news.ycombinator.com/item?id=46936870

Points: 1

# Comments: 0

Article URL: https://thejaggi.blogspot.com/2026/02/beyond-wapo-angst-why-journalists-need.html

Comments URL: https://news.ycombinator.com/item?id=46936862

Points: 1

# Comments: 0

I switched from Linux to macOS a while ago and there are two things I still miss:

Launching apps with parameters. On Linux you edit a .desktop file and that's it — VS Code opening a project folder, a script, whatever. On macOS there's no simple way to do this. Automator and Shortcuts feel overcomplicated for something so basic, and the paid launchers do way more than I need. Alt+Tab showing windows, not apps. Cmd+Tab groups everything by application. If I have several Chrome or VS Code windows open, I can't pick which one I want. So I vibecoded this. It does two things: - Cmd+Shift+Space opens a grid of configurable launchers. Each one runs a shell command — open an app, a URL, a script, with whatever arguments. - Option+Tab shows every window individually, like Alt+Tab on Linux. It's a native Swift app, JSON config, no dependencies. Free and open source.

Comments URL: https://news.ycombinator.com/item?id=46936845

Points: 1

# Comments: 0

Article URL: https://blog.svgames.pl/article/the-little-bool-of-doom

Comments URL: https://news.ycombinator.com/item?id=46936828

Points: 2

# Comments: 0

Article URL: https://www.nature.com/articles/s41598-026-36004-4

Comments URL: https://news.ycombinator.com/item?id=46936819

Points: 1

# Comments: 0

Article URL: https://isledb.com/

Comments URL: https://news.ycombinator.com/item?id=46936813

Points: 1

# Comments: 0

Article URL: https://www.columbialawreview.org/content/on-algorithmic-wage-discrimination/

Comments URL: https://news.ycombinator.com/item?id=46936805

Points: 1

# Comments: 0

Article URL: https://gitbutler.com/

Comments URL: https://news.ycombinator.com/item?id=46936803

Points: 1

# Comments: 0

Article URL: https://bcbusiness.ca/industries/general/canadian-startups-stop-playing-slow/

Comments URL: https://news.ycombinator.com/item?id=46936800

Points: 1

# Comments: 0

Article URL: https://httpsornot.com/

Comments URL: https://news.ycombinator.com/item?id=46936772

Points: 1

# Comments: 1

Article URL: https://goodinternetmagazine.com/web-design-without-design-software/

Comments URL: https://news.ycombinator.com/item?id=46936765

Points: 2

# Comments: 0

Article URL: https://github.com/ClassicOldSong/Plexsonic

Comments URL: https://news.ycombinator.com/item?id=46936734

Points: 1

# Comments: 0

Article URL: https://www.politico.eu/article/council-and-parliament-agree-to-delay-and-review-eu-deforestation-law/

Comments URL: https://news.ycombinator.com/item?id=46936732

Points: 2

# Comments: 0

Hey HN!

I’ve been working on Obelis AI — it’s like having a DevOps engineer on your team, powered by AI agents (real ones, not just a fancy LLM wrapper).

The idea came from seeing how early-stage teams often push off infra decisions until it’s too late. Migrations are always an option, but they get painful fast as things scale.

Platforms like Vercel and Firebase are great to get going, but they tend to get expensive, and you don’t really own the infra. We try to give you the same simplicity — but everything runs on your own cloud (AWS for the timebeing), with full control.

We would love feedback, questions, or any brutally honest thoughts. Appreciate it!

Comments URL: https://news.ycombinator.com/item?id=46936725

Points: 1

# Comments: 0